Malignant gliomas (MGs) are univerally fatal, and effective therapy is limited by collateral damage to normal tissue. Immunotherapy directed against tumor-specific antigens may allow neoplastic cells to be targeted more precisely, and our dendritic cell (DC)-based vaccinations targeting of a mutated tumor-specific epidermal growth factor receptor have produced immunologic and radiographic responses in patients with MGs. The discovery that MGs, but not surrounding normal brain, serve as a refuge for Cytomegalovirus (CMV) reactivation provides an unparalleled opportunity to subvert, as a tumor-specific antigen, the highly immunogenic CMV protein, pp65. Despite the numerous advantages of targeting CMV antigens in MGs with DC-based vaccines, a number of factors clearly limit ant/tumor immune responses in these patients. Innovative complementary strategies that eliminate CD25+ regulatory T cells or block cytotoxic 3; lymphocyte antigen-4-induced T cell tolerance may enhance such immune responses, but the indiscriminate application of these potent adjuvants carries the risk of inducing autoimmune encephalomyelitis. In order to understand the limitations and risks of targeting CMV antigens in MGs, we have developed a novel murine astrocytoma cell line that supports infection with murine CMV and is tumorigenic in syngeneic mice. Our preliminary murine studies demonstrate that these tumors in the brain can be targeted with RNA-loaded DCs. We have also shown that DCs from patients with MGs that are loaded with pp65mRNA, induce interferon-gamma, production from CD4+ and CDS+ T-cells in an antigen-specific manner and incite T-cells to kill malignant astrocytes infected with human CMV. Interestingly, we have also found that CMV-specific T-cells preferentially accumulate at the tumor site in patients with MGs. We believe that our murine model system and the complementary human studies proposed will allow selection and translation of the most effective strategies for targeting CMV-associated antigens in patients with MGs, without the induction of autoimmunity. In this project, we will use the murine model, in combination with in vitro human studies to evaluate the safety of, and to gain a better understanding of the mechanisms involved in the therapeutic targeting of CMV-associated proteins in malignant gliomas. The results will then be used to rationally design and conduct a clinical CMV-targeted clinical trial.